Recent questions tagged machine-learning

0 0 votes
0 0 answers
101
101 views
Suppose our true labels are $\vec{y}=[0,0,1]$, our predicted probabilities of being in class 1 are $[0.1,0.6,0.9]$, and our threshold is $T=0.5$. Give the total (not aver...
0 0 votes
0 0 answers
164
164 views
Which of the following statements about logistic regression are correct? The cost function of logistic regression is convex.The cost function of logistic regression is co...
0 0 votes
1 1 answer
192
192 views
$\text { Which of the following functions is the logistic loss for label } y=+1 \text { ? }$ABCD
1 1 vote
1 1 answer
150
150 views
Select always, sometimes, or never to describe when each statement below is true about a logistic regression model $P(Y=1 \mid X)=\sigma\left(X^T \beta\right)$, where $Y$...
0 0 votes
0 0 answers
117
117 views
What is the value of the sigmoid function $\sigma(z)=\frac{1}{1+e^{-z}}$ when $z=2 ?$0.1190.2680.50.881
0 0 votes
1 1 answer
145
145 views
Mark the expression that describes the odds ratio $\frac{P(Y=1 \mid X)}{P(Y=0 \mid X)}$ of a logistic regression model.Recall: $P(Y=0 \mid X)+P(Y=1 \mid X)=1$ for any $X$...
0 0 votes
1 1 answer
155
155 views
Mark the expression that describes $P(Y=0 \mid X)$ for a logistic regression model. $\sigma\left(-X^T \beta\right)$ $1-\log \left(1+\exp \left(X^T \beta\right)\right)$ $1...
0 0 votes
1 1 answer
216
216 views
What is the purpose of the sigmoid function in logistic regression ? It converts continuous input into categorical data.It standardizes the input to have zero mean and va...
0 0 votes
1 1 answer
236
236 views
In which of the following situations can logistic regression be used? Select all that apply. Predicting whether an email is a spam email or not based on its contents.Pred...
0 0 votes
1 1 answer
170
170 views
$\text { Logistic regression is actually used for classification. }$(Please enter 1 for True and 0 for False). 
1 1 vote
1 1 answer
501
501 views
Consider linear regression and logistic regression. They both use linear functions.They both can be used to solve regression prob-They both use the logistic activation fu...
0 0 votes
0 0 answers
109
109 views
Suppose we trained a logistic regression classifier for some binary classification task. The true labels $y$ and predicted probabilities $P(Y = 1|x)$ are given below.$$y$...
0 0 votes
0 0 answers
93
93 views
Consider the dataset shown in the diagram below with a logistic regression decision boundary.Questions :1. Count the number of mistakes the model is making.2 . Is the dat...
0 0 votes
1 1 answer
162
162 views
Suppose that you have trained a logistic regression classifier $h_{\theta}(x) = \sigma(1 - x)$ where $\sigma(.)$ is the logistic/sigmoid function. What does its output on...
0 0 votes
1 1 answer
110
110 views
Logistic regression sends an example $x$ through a linear function to get a real number, which it sends through an exponential function to get a positive number, which it...
0 0 votes
1 1 answer
180
180 views
Consider a logistic regression model with $w = (1, 2)$ and $b = 0$. Use the model to estimate probability that $y$ is $1$, given that $x$ is $(-3, 1)$.0.2690.006690.00247...
0 0 votes
1 1 answer
177
177 views
Suppose you are given the following classification task: predict the target $Y \in \{0, 1\}$ given two real valued features $X_1 \in \mathbb{R}$ and $X_2 \in \mathbb{R}$ ...
0 0 votes
0 0 answers
102
102 views
If $\theta = \begin{bmatrix}1 \\ 4 \\ 3\end{bmatrix}$ and $x = \begin{bmatrix}7 \\ -1 \\ -1\end{bmatrix}$, what is the value of $P(y = 1)$?00.250.51
1 1 vote
2 2 answers
211
211 views
What is the range of the sigmoid function?$ (0, 1 )$$(-1, 1)$$(-\infty, -\infty)$$(0, -\infty)$
0 0 votes
2 2 answers
235
235 views
What is the value of the sigmoid function $\sigma(x) = \frac{1}{1 + e^{-x}}$ when $x = 2$ ? 0.119 0.268 0.5 0.881 
0 0 votes
1 1 answer
195
195 views
Suppose you're using $L2$ regularization on a least squares objective. Some value $\lambda^*$ will give you the best test error among all possible $\lambda$. You train yo...
0 0 votes
1 1 answer
216
216 views
We are solving a least-squares linear regression problem without regularization. Suppose that the following twoweight matrices both have the same cost: $W1 = \begin{bmatr...
0 0 votes
1 1 answer
142
142 views
Which of the following statements are true about Lasso and ridge regression?Both ridge regression and Lasso are methods used to reduce overfitting that might occur in sta...
0 0 votes
0 0 answers
125
125 views
Ridge regression can shrink all coefficients to exactly O if the regularization parameter $\lambda$ is large enough.Please enter 1 for True and 0 for False.
0 0 votes
0 0 answers
106
106 views
is $X^TX + \lambda I$ invertible for $ \lambda 0$?AlwaysIf $\lambda$ is larger than the smallest eigenvalue of $X^T X$If $X$ is full rankNever
0 0 votes
0 0 answers
129
129 views
Given normal equation : $X^T X w = X^Ty$If $X^T X$ is not invertible, do the normal equations still define the solution?Please enter 1 for True and 0 for False.
1 1 vote
0 0 answers
147
147 views
suppose Alice and Bob do the same experiment • Alice measures distance in mm • Bob measures distance in kmthey each compute an estimator with ridge regression and c...
0 0 votes
0 0 answers
145
145 views
suppose Alice and Bob do the same experiment • Alice measures distance in mm • Bob measures distance in kmthey each compute an estimator with least squares and comp...
0 0 votes
1 1 answer
126
126 views
In a LASSO Regression, if the regularization parameter $\lambda$ is very high, which of the following is true? The model can shrink the coefficients of uninformative feat...